Steric Hindrance By APRIL in Binding of BCMA-Directed Bispecific Antibodies: A Novel Refractory Mechanism of BCMA-Directed Immunotherapies in Multiple Myeloma

Multiple myeloma (MM) cells preferentially reside and grow in the bone marrow while activating osteoclasts (OCs). The aberrantly activated OCs induce bone lysis and support MM-cell growth and survival, leading to forming the vicious cycle between MM progression and bone destruction. B-cell maturation antigen (BCMA) is expressed in terminally differentiated plasma cells and MM cells, thereby drawing attention as a therapeutic target in MM. Indeed, BCMA-directed immunotherapies have been recently utilized as the standard of care against refractory and/or relapsed MM (RRMM). Despite their superb initial response against RRMM, MM eventually relapses due to various mechanisms, including target loss/mutation and immune impairment. However, the contribution of OCs to MM relapse is largely unknown in the setting of BCMA-directed immunotherapies. The present study aimed to clarify the roles of OCs in efficacy of BCMA-directed therapies. Since OCs robustly secret BCMA ligands, a proliferation-inducing ligand (APRIL) and BAFF, we first examined whether APRIL and/or BAFF affect the binding of anti-BCMA antibodies to BCMA on MM cells. After incubation with APRIL but not BAFF, membrane-bound BCMA (mBCMA) on MM cells became undetectable in FCM using anti-BCMA monoclonal (clone: 19F2 and REA315) and polyclonal antibodies (Abs). However, the APRIL treatment did not affect BCMA expression in MM cells at protein and mRNA levels using immunoblotting and quantitative PCR, respectively. Interestingly, surface binding of biotinylated APRIL on MM cells was observed in FCM; mBCMA was not internalized into cytoplasm in MM cells upon treatment with APRIL in FCM, whereas endocytosis of CXCR4 was induced in MM cells by SDF-1. The APRIL treatment upregulated NF-kB-related transcripts without affecting the BCMA (TNFRSF17) transcript in MM.1S and H929 cells using RNA-seq and gene set enrichment analysis. mBCMA is shed off as soluble BCMA (sBCMA) by γ-secretase; treatment with the γ-secretase inhibitor RO4929097 substantially increased mBCMA levels on MM cells while reducing sBCMA. However, APRIL pre-treatment similarly impaired the detection of mBCMA on MM cells in FCM without affecting sBCMA levels in the presence of RO4929097. Because APRIL can bind to the heparan sulfate proteoglycan CD138, we next examined the roles of CD138 in the impairment of mBCMA detection on MM cells in FCM in the presence of APRIL. We generated CD138 (SDC1)-knockout RPMI 8226 cells using the lentiviral CRISPR-Cas9 system. In the SDC1-knockout RPMI 8226 cells, reduction of mBCMA levels in FCM was minimized and degradation of BCMA total protein appeared to be faster in cycloheximide chase assay in the presence of APRIL compared to those in parental RPMI 8226 cells. These results suggest that APRIL hampers binding of anti-BCMA antibodies without affecting BCMA production in MM cells. We further investigated the effects of APRIL on binding of the anti-BCMA/anti-CD3 bispecific Abs, teclistamab and elranatamab to BCMA in MM cells. APRIL pretreatment impaired the binding of these bispecific Abs to BCMA on MM cells in a dose-dependent fashion. Importantly, cytotoxic activity of these bispecific Abs against MM cells was impaired in the presence of APRIL, accordingly. OCs are a major source of APRIL in MM bone marrow, and support MM cell growth and survival. The present results demonstrate that APRIL hinders the binding of antibodies to BCMA on MM cells, including therapeutic bispecifics, teclistamab and elranatamab. APRIL may also affect mBCMA turnover in a manner dependent on CD138. Therefore, APRIL derived from OCs might cause steric hindrance of BCMA from anti-BCMA Ab binding and alleviate the efficacy of immunotherapies targeting BCMA. Further study is warranted to overcome the resistant mechanisms for BCMA-directed immunotherapies.